Green cryptocurrencies are supposed to hedge clean-energy equity, but the paper gives traders no hedge ratio, portfolio weight or hedge-effectiveness figure. The conclusion makes the claim. Beyond descriptive statistics and a correlation matrix, the measured results are variance decompositions. Those tables deserve attention, though about a quarter of their cross-spillover mass comes from three near-identical ESG indices exchanging shocks.

The ten-series network

The authors use daily returns from January 2018 to December 2024 for ten series:

They keep crypto weekdays only, aligning it with bonds and equities. A TVP-VAR lets autoregressive coefficients drift as a random walk in place of a rolling window. The authors use its generalized forecast-error variance decomposition at a 10-day horizon. Each cell assigns a share of one series' forecast-error variance to shocks from another. The total connectedness index (TCI) averages the off-diagonal shares; TO, FROM and NET identify transmitters and receivers. We did not find a definition of the volatility series, or an observation count.

Full-sample TCI reaches 62.10% for returns and 59.58% for volatility. During the pre-COVID years, 2018-2019, connectedness ran at 55-65%. In early 2020, volatility TCI peaked above 85% and return TCI reached almost 80%. The paper considers two global crises: COVID-19 (2020-2021) and the Russia-Ukraine war, labelled 2022 in one figure and 2022-2023 in another. The abstract takes the crisis jump as evidence that "global shocks heighten market interdependence." After the pandemic, volatility TCI fell to nearly 50%; return TCI held at 55-60%.

The ESG pillars lead net volatility transmission at +16.61, +16 and +15.96. ADA (+5.63) is the only coin sending more volatility than it receives. Green bonds (-21.01), gold (-15.1) and clean energy (-8.99) are net receivers.

The extension over earlier work is narrow. Hypothesis H4 anticipated the crisis jump, and the authors cite Elsayed et al. (2022a) for a higher COVID TCI. Yousaf et al. (2024) and Alharbi et al. (2025), also cited here, had already compared green cryptocurrencies with ESG stock indices. Both found receiving coins and transmitting green stocks. This paper puts all ten series into one TVP-VAR system.

How much of the hub is repeated equity risk?

The main network result casts ESG equities as net transmitters of returns and volatility. The volatility table's block of ESG-to-ESG flows makes that result harder to interpret. Each pillar draws 23.65 to 23.94 points from each sibling. Environment's FROM is 75.72, including 47.48 from Governance and Social; its TO is 92.33, including 47.55 sent to them.

Across the six intra-ESG cells, those flows total 142.86 of 595.84 cross-spillover points, or 24% of the network. The return table has 152.41 of 620.98 in the same block. A crude subtraction, without re-estimating the model, takes volatility TCI from 59.58 to roughly 45.

The NET signs survive that subtraction. Environment sends about 44.8 points to non-ESG assets and receives about 28.2. The size of the wider spillover is less certain: together, the three pillars supply 40.0 of clean energy's volatility FROM of 58.19 and 34.05 of green bonds' 54.17. A generalized decomposition gives three copies of one equity shock three opportunities to claim variance. We did not find a run with a single ESG node that could separate equity risk from triplication.

Timing raises another question. Environment's Ljung-Box Q20 is 85.80, against Bitcoin's 26.01. The pillars' excess kurtosis, 13.94 to 15.15, resembles Bitcoin's 14.23. Such strong serial correlation in a global equity index is consistent with mixed closing times. A VAR can read the resulting lead-lag as spillover. We did not find the time at which each series is stamped.

Receiver status and a hedge

The paper describes gold, green bonds and clean energy as receivers, "underlining their defensive nature and potential role as hedging instruments during periods of turmoil." Its equation 3 accumulates squared impulse responses, leaving spillover cells without a sign. A receiver that falls alongside a transmitter prints the same number as one that rises against it.

Row-by-row normalization matters too. Green bonds' return FROM of 52.92 is a share of daily return variance of 0.1574; clean energy's variance is 2.7484. A large share of a small variance can still be a small absolute amount. We raised a related concern about low green-bond volatility driving portfolio results in an earlier note. The assets called defensive also transmit to each other. Gold's 11.59 accounts for 22% of green bonds' return FROM of 52.92. Green bonds' 13.86 accounts for 42% of gold's FROM of 32.8.

Bitcoin gets conflicting descriptions. The abstract and conclusion say it becomes a volatility receiver in crises. Discussing the dynamic plot, the paper says it "turns into net volatility transmitter" during COVID-19. Its full-sample volatility NET is -1.34.

Hedge methods, absent hedge results

The methods are there. The paper defines dynamic weights after Kroner and Ng (1998), a hedge ratio after Kroner and Sultan (1993), and hedge effectiveness after Ku et al. (2007). The results proceed through a descriptive table, two spillover tables and seven figures before ending on the TCI plot. We found no reported weight, ratio or effectiveness value.

Hypothesis H5 asks whether "Green cryptocurrencies provide a hedging function for clean energy index." The conclusion answers without reporting that test. Green cryptocurrencies "demonstrate superior performance as hedging instruments for clean energy assets, offering high hedge effectiveness and consistent optimal weighting across various market regimes," it says. Gold and green bonds receive different assessments. The abstract grants them a "potential role as hedging instruments during periods of turmoil," whereas the conclusion calls them "relatively inefficient hedging tools in the context of clean energy exposure." Neither assessment has a figure attached.

The published inputs suggest a low ceiling for the claim. Take the paper's 0.32 correlation between clean energy and ADA at face value. A static minimum-variance hedge removes the squared correlation, about 10% of variance. Given ADA's daily variance of 34.7554, the implied hedge ratio is roughly 0.09 units of ADA per unit of clean energy. A dynamic ratio could do better, though it would need to remove far more than the static 10% to warrant "high".

There is also a mismatch between the correlation matrix and spillover tables. Figure 2 shows an ESG series correlated at 0.72 with ADA, about 52% when squared. The text calls it the S&P Global ESG Index, which is outside the model's three STOXX pillars. Those pillars together explain 6.22 points of ADA's return variance. Bitcoin and ADA have a correlation of -0.26, under 7% when squared, while ADA explains 19.22 points of Bitcoin's return variance. The text quotes 18.32, the IOTA cell.

A table showing hedge effectiveness for clean energy against each coin, split across the authors' three periods and net of costs, would change our view.

The universe we could not rebuild

We have no price data for the S&P Green Bond Index, and IOTA falls outside our crypto coverage. Substituting US-listed green-bond, ESG, clean-energy and gold ETFs plus spot BTC, ADA and XRP would produce a different network. We could attribute none of its relationships to the paper's universe. Nor could we backtest the authors' hedge: they publish no weights or ratios.